5.4.1. Packer Fluids: Wang Et Al., 2005 Vollmer Et Al. (2004)

Tubing Well Performance, Heat Transfer and Sizing 283
Maintaining cold injection temperatures to aid in thermal fracturing of the

formation in water injection wells
Flow assurance (wax, hydrate, asphaltenes, etc.) for deepwater completions
All of these issues require reducing the heat transfer. Occasionally, increasing the
heat transfer away from the produced fluids is beneficial. One example is with high-
temperature well tests where there are temperature limitations on surface equipment
such as BOPs.
A number of insulating methods have been applied over the years.
5.4.1. Packer fluids

Not all wells use a sealed annulus, but those that do can pump a fluid into the
annulus above the packer or in the riser. The method of deploying the fluid is either
to use a sliding sleeve or, if this potential leak path is to be avoided, to forward or
reverse circulate the fluid into the annulus once the tubing hanger has been set but
prior to setting a packer.
To be a good insulator, the fluid should have a low conductivity and also limit-
free convection. With a water or brine packer fluid, convection will be turbulent.
Oils appear to be an excellent choice due to their low conductivity (around
0.08 Btu/h ft 1F), but lower conductivities lead to higher temperature gradients, and
heat transfer via convection may be 10–20 times that of conduction (Wang
et al., 2005). Interestingly a method described by Vollmer et al. (2004) for
decreasing the wellbore flowing temperature is to deploy friction reducers with the
packer fluid. Going in the opposite direction by increasing the viscosity – initially
with diesel and now with other viscosified mineral oils – is routine. Environmental
issues may limit the use of oil-based systems. Water-based systems, sometimes with
glycol for reduced conductivities (Dzialowski et al., 2003), are now increasingly
used, but conductivities of oil are still lower than that of water. Regardless of the
fluid, it has to be thermally stable and the yield point or the viscosity of the fluids has
to be high. Wang et al. (2006) report significant reductions in heat transfer with
viscosities as high as 11,000 cp, whilst Horton et al. (2005) propose a high yield
stress (10–105 lbf/100 ft2) solution. Modelling of these, often non-Newtonian,
fluids, has to be precise at low shear rates in order for simulations to be accurate.
Gas makes an excellent insulator at low pressure. Nitrogen is commonly used.
The conductivity (and convection) of a gas is directly related to its density and
hence to its pressure. Gas lift, for example, will have a marginal impact on the
effective conductivity due to the relatively high pressures. The annular flow of lift
gas will, however, have a marked effect close to the wellhead. The relatively low
heat capacity of the gas means that it heats up (or cools down) quickly as it flows
down the annulus. A further cooling effect is observed due to Joule–Thomson
cooling both at the gas lift valve and within the commingled flow stream in the
tubing.
For an insulation effect, the gas first has to be circulated down the annulus. The
annulus then has to be sealed (e.g. a packer set) and the gas pressure released. For
leak detection purposes, a low pressure is often left on the annulus (or riser) but
284 Temperature Control
Temperature (°F)
220 240 260 280 300
0
2000
4000
Depth (ft, TVD)
Tubing: water filled annulus

Tubing: gas filled annulus 3000 psig
6000 Tubing: gas filled annulus 1000 psig
8000 Tubing: gas filled annulus 0 psig
Production casing: water filled annulus
Production casing: gas filled annulus 3000 psig
Depth of gas filled annulus
10000 Production casing: gas filled annulus 500 psig
12000
Figure 5.22 Gas-¢lled annulus e¡ect.
even this will have deleterious consequences. Figure 5.22 shows the modelled effect
of a nitrogen-filled annulus on the tubing and production casing temperatures. The
models are for steady-state production of a mixture of oil, water and gas. Nitrogen is
not the most effective gas, although it is cheapest. Argon can be used (e.g. as used
for inflation of deepwater diving suits), but is unlikely to be cost-effective, mainly
because far more gas is required to displace the liquid than is ultimately required,
thus wasting large quantities or requiring gas capture and storage.
These low pressures can cause problems such as casing collapse. They could also
promote fluid ingress which could go undetected. The hot production and low
pressures can also cause vaporisation of underlying packer fluids (if present), leading
to reduced conductivities through refluxing. This effect was first observed in steam
injection wells (Aeschliman, 1985). Where conduction from non-insulated
connections causes packer fluids to vaporise, the resulting steam condenses on the
casing walls higher up the well. The water then drops down the well, and
the process repeats. Even with no gauge pressure on the annulus, and most of the
annulus full of super-heated steam, this heat transfer loop can continue undetected
(Willhite and Griston, 1987). Away from steam injection wells, this effect can still
occur even on relatively cool wells, especially where the liquid level and pressures
are low. It is also a potential corrosion mechanism (Chapter 8), especially in HPHT
wells. Therefore, the predictions shown in Figure 5.22 should be treated with
caution – Aeschliman et al. (1983) report heat losses three to six times greater in a
steam injection well due to this effect. An improved method can be envisaged
through over-displacing the annulus to gas, setting a production packer and then
Tubing Well Performance, Heat Transfer and Sizing 285
applying and maintaining a near vacuum. Maintaining a vacuum also lowers the
boiling point of the packer fluid. Many connections and components are not
designed to be leak proof against a vacuum.
Silicate foams and gels such as aerogels are widely reported and used in
construction. They have a low density, but are strong and stable for their weight and
can be pumped into a well if somehow kept dry. They can be created in situ
(Kuperus et al., 2001), but this does require several steps including critical point
drying by circulating carbon dioxide. If successfully deployed, they can have
effective conductivities as low as 0.01 Btu/h ft 1F.
Practical considerations for all potential packer fluids include compatibility with
control line encapsulation, crystallisation at low temperatures, corrosion potential,
compatibility with elastomers, other completion fluids ( Javora et al., 2002) and the
potential for (and effect of ) contamination, for example from control line contents
or by mixing in rig tanks. Much of the advantage of a low-conductivity, low-
convection fluid will also be lost if there is tubing to casing contact. This contact is
inevitable due to buckling or deviation in all wells unless centralisers are used. The
centralisers should be spaced according to the frequency of the buckling (Section
9.4.8, Chapter 9). In wells with control lines and gauge cables, centralisers can be
incorporated into cable clamps. Care must be taken to ensure that the centralisers
are robust to buckling and installation loads and incorporate insulation (such as
plastic pads) to prevent indirect metal-to-metal contact between the tubing and the
casing.
5.4.2. Low-density cements

Low-density cements (e.g. foamed or incorporating microspheres (Rae and Lullo,
2004) are now routinely deployed in deepwater completions (Benge and Poole,
2005; Piot et al., 2001) and for other low fracture gradient applications (Ravi
et al., 2006). A by-product of the low density of these cements is lower thermal
conductivity. Glass G cements have conductivities around 0.9 Btu/h ft 1F. Low-
density cements may have conductivities at or below 0.4 Btu/h ft 1F. These then
provide an easy opportunity for insulation – above all because they may be deployed
anyway in the critical area close to the mudline of a deepwater well, for example. By
using a larger hole size or a longer cement column, further insulation may be
obtained.
5.4.3. Thin-film insulation

Thin-film insulation is a modified liquid epoxy coating (Lively, 2002) that can be
applied in multiple, very thin layers to the outside of tubing and casing. The
multiple layers may only add 0.25 in. to the outside diameter and hence can be used
in tighter clearance wells than, for example, VIT. The conductivity of the thin films
may be as low as 0.04 Btu/h ft 1F. Because of their slenderness they are best used in
association with an insulating packer fluid.

5.4.1. Packer Fluids: Wang Et Al., 2005 Vollmer Et Al. (2004)

Uploaded by

Copyright:

Available Formats

5.4.1. Packer Fluids: Wang Et Al., 2005 Vollmer Et Al. (2004)

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

5.4.1. Packer Fluids: Wang Et Al., 2005 Vollmer Et Al. (2004)

Uploaded by

Copyright:

Available Formats

Tubing Well Performance, Heat Transfer and Sizing 283

Maintaining cold injection temperatures to aid in thermal fracturing of the